Tongliang Liu

Lecturer (Assistant Professor)
School of Computer Science
Faculty of Engineering and Information Technologies
The University of Sydney

Email: tliangDOTliuATgmailDOTcom

Postal Address:
Room 315, Level 3, J12
Cleveland St, Darlington NSW 2008, Australia


CV    Research    Publications    Google Scholar


About me: 

Greetings! I am currently a Lecturer with the School of Information Technologies at The University of Sydney. I received the PhD degree from the University of Technology Sydney and the BE degree in Electronic Engineering and Information Science from the University of Science and Technology of China. Before joining USyd, I was a Lecturer with the Centre for Artificial Intelligence and the School of Software at the University of Technology Sydney.

My research interests lie in providing mathematical and theoretical foundations to justify and further understand machine learning models and designing efficient learning algorithms for problems in computer vision and data mining, with a particular emphasis on matrix factorisation, transfer learning, multi-task learning, and learning with label noise.

I am always looking for highly-motivated Mphil/MSc and PhD students to join our group. If you are interested in the research in our group, please send an email to tongliang.liu@sydney.edu.au about your interests and background (attaching your CV, transcripts, and any previous research papers). Thanks!


News: 

  • 09/2018, our paper on multi-task learning has been accepted by IEEE TNNLS.
  • 08/2018, I accepted the invitations to serve as a reviewer for (NIPS2018, currently serving) ICLR 2019, AISTAS2019; and as a Senior program committee for AAAI2019.
  • 07/2018, our papers on complementary label, domain generalization, and adapted triplet loss have been accepted by ECCV 2018. Congrats! Xiyu! Ya! and Baosheng!
  • 06/2018, our paper titled Deep Blur Mapping: Exploiting High-Level Semantics by Deep Neural Networks has been accepted by T-IP.
  • 04/2018, we have theoretically answered why "going deeper, generalising better" and why deep learning algorithm can have a small test error. Details are in the arXiv paper.
  • 04/2018, our papers on quantum NMF, transfer learning, and hashing have been accepted by IJCAI 2018. Congrats! Yuxuan! Yong! and Erkun!
  • 03/2018, our paper titled Transferring Knowledge Fragments for Learning Distance Metric from A Heterogeneous Domain has been accepted by T-PAMI.
  • 03/2018, our paper titled An Efficient and Provable Approach for Mixture Proportion Estimation Using Linear Independence Assumption has been accepted by CVPR. Congrats! XiYu!

  •  

    Last Update: 9/2018.